基于文本分类的电企舆情识别方案设计与实现

摘要

According to the need of electric power enterprise management and control of public opinion , the experimental platform of public opinion was designed based on text classification , technicians set different parameters in text classification and test effect of different parameters , finally this paper used JAVA to realize the theme of public opinion system under the specific program. Practical application shows that this platform and system for the enterprise , further deepening the control provides reliable public opinion research platforms and application systems , thereby raising the standard of corporate brand building and brand maintenance.%根据电力企业舆情管控工作的需要,设计了一种基于文本分类技术的企业舆情主题识别实验平台,技术人员只需根据需要设定文本分类中不同的基本参数,即可针对相应的舆情文本测试不同参数方案下主题文本的识别效果,最后利用JAVA实现了特定方案下舆情主题识别系统(原型).实际应用表明,该实验平台和原型系统能够为本企业进一步深化舆情管控工作提供可靠的研究平台和应用系统,从而提升企业品牌建设和品牌维护工作的水平.